# Membership Inference Attack for Beluga Whales Discrimination

## TL;DR

This paper introduces a novel application of Membership Inference Attacks to discriminate between known and unknown beluga whales in digital photos, aiding wildlife monitoring and re-identification efforts.

## Contribution

The study applies MIAs to animal ecology, demonstrating their effectiveness in whale discrimination and proposing a new ensemble MIA strategy to improve accuracy and reduce false positives.

## Key findings

- MIA-based approach achieves high discrimination accuracy on whale datasets.
- Ensemble MIA improves attack performance over individual MIAs.
- Method leverages privacy attack techniques for ecological research benefits.

## Abstract

To efficiently monitor the growth and evolution of a particular wildlife population, one of the main fundamental challenges to address in animal ecology is the re-identification of individuals that have been previously encountered but also the discrimination between known and unknown individuals (the so-called "open-set problem"), which is the first step to realize before re-identification. In particular, in this work, we are interested in the discrimination within digital photos of beluga whales, which are known to be among the most challenging marine species to discriminate due to their lack of distinctive features. To tackle this problem, we propose a novel approach based on the use of Membership Inference Attacks (MIAs), which are normally used to assess the privacy risks associated with releasing a particular machine learning model. More precisely, we demonstrate that the problem of discriminating between known and unknown individuals can be solved efficiently using state-of-the-art approaches for MIAs. Extensive experiments on three benchmark datasets related to whales, two different neural network architectures, and three MIA clearly demonstrate the performance of the approach. In addition, we have also designed a novel MIA strategy that we coined as ensemble MIA, which combines the outputs of different MIAs to increase the attack accuracy while diminishing the false positive rate. Overall, one of our main objectives is also to show that the research on privacy attacks can also be leveraged "for good" by helping to address practical challenges encountered in animal ecology.